Interface BuiltInAdaptiveVadProviderConfig

Configuration for the BuiltInAdaptiveVadProvider.

All fields are optional — sensible defaults are applied for 16kHz mono audio with 20ms frames (the standard voice pipeline configuration).

See

BuiltInAdaptiveVadProvider for usage examples See AdaptiveVADConfig for detailed VAD tuning parameters. See CalibrationConfig for environmental calibration settings.

interface BuiltInAdaptiveVadProviderConfig {
    sampleRate?: number;
    frameDurationMs?: number;
    calibration?: CalibrationConfig;
    vad?: AdaptiveVADConfig;
}

Properties

sampleRate? frameDurationMs? calibration? vad?

Properties

Optional sampleRate

sampleRate?: number

Audio sample rate in Hz. Must match the actual audio stream.

Default

16_000 (16 kHz, standard for voice processing)

Optional frameDurationMs

frameDurationMs?: number

Duration of each audio frame in milliseconds. Determines the number of samples expected in each processFrame() call. At 16kHz with 20ms frames, each frame is 320 samples.

Default

20

Optional calibration

calibration?: CalibrationConfig

Environmental calibration settings. Controls how the noise floor and spectral profile are estimated from ambient audio.

See CalibrationConfig for available options.

Optional vad

vad?: AdaptiveVADConfig

VAD algorithm configuration. Controls speech detection thresholds, minimum speech duration, hangover period, etc.

See AdaptiveVADConfig for available options.

Settings

Member Visibility

Theme

On This Page

sampleRateframeDurationMscalibrationvad